options(knitr.duplicate.label = 'allow')
DATA_PATH <- here("data/processed/syntactic_bootstrapping_tidy_data.csv") # make all variables (i.e. things that might change) as capital letters at the top of the scripts
ma_data <- read_csv(DATA_PATH)
ma_data <- ma_data %>% filter(paradigm_type != "action_matching")
n_effect_sizes <- ma_data %>%
filter(!is.na(d_calc)) %>%
nrow()
n_papers <- ma_data %>%
distinct(unique_id) %>%
nrow()
There are 16 effect sizes collected from 4 different papers.
Here are the papers in this analysis:
ma_data %>%
count(short_cite) %>%
arrange(-n) %>%
DT::datatable()
## Forest plot
ma_model <- rma(ma_data$d_calc, ma_data$d_var_calc)
ma_model
##
## Random-Effects Model (k = 16; tau^2 estimator: REML)
##
## tau^2 (estimated amount of total heterogeneity): 6.1524 (SE = 2.3713)
## tau (square root of estimated tau^2 value): 2.4804
## I^2 (total heterogeneity / total variability): 97.75%
## H^2 (total variability / sampling variability): 44.46
##
## Test for Heterogeneity:
## Q(df = 15) = 218.5173, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## 0.6081 0.6386 0.9523 0.3410 -0.6435 1.8597
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
forest(ma_model,
header = T,
slab = ma_data$unique_id,
col = "red",
cex = .7)
## funnel plot {.tabset} ### vanilla
ma_data %>%
mutate(color = ifelse(sentence_structure == "transitive", "red", "blue"),
color_sample_size = ifelse(n_1 < 10, "red", ifelse(n_1 < 20, "orange", "yellow")),
color_confidence = ifelse(inclusion_certainty == 1, "red", "black"))-> ma_data_funnel
ma_data_funnel %>% filter (abs(d_calc) < 5) -> ma_data_funnel_no_outlier
ss_colors <- ma_data_funnel$color
ss_colors_no_outlier <- ma_data_funnel_no_outlier$color
ma_model_funnel <- rma(ma_data_funnel$d_calc, ma_data_funnel$d_var_calc)
ma_model_funnel_no_outlier <- rma(ma_data_funnel_no_outlier$d_calc, ma_data_funnel_no_outlier$d_var_calc)
f1<- funnel(ma_model_funnel, xlab = "Effect Size", col = ss_colors)
legend("topright",bg = "white",legend = c("transitive","intransitive"),pch=16,col=c("red", "blue"))
title(main = "All effect sizes break down by sentence structure")
f2<- funnel(ma_model_funnel_no_outlier, xlab = "Effect Size", col = ss_colors_no_outlier)
legend("topright",bg = "white",legend = c("transitive","intransitive"),pch=16,col=c("red", "blue"))
title(main = "effect sizes excluded outliers (abs <5) break down by sentence structure")
ma_model_sentence_structure <- rma(ma_data_funnel$d_calc~ma_data_funnel$sentence_structure, ma_data_funnel$d_var_calc)
ma_model_no_outlier_ss <- rma(ma_data_funnel_no_outlier$d_calc~ma_data_funnel_no_outlier$sentence_structure, ma_data_funnel_no_outlier$d_var_calc)
f3 <- funnel(ma_model_sentence_structure, xlab = "effect size", col = ss_colors)
f3_b <- funnel(ma_model_no_outlier_ss, xlab = "effect size", col = ss_colors_no_outlier)
ma_model_ss_age <- rma(ma_data_funnel$d_calc~ma_data_funnel$sentence_structure + ma_data_funnel $mean_age, ma_data_funnel$d_var_calc)
ma_model_funnel_no_outlier_ss_age <- rma(ma_data_funnel_no_outlier$d_calc~ma_data_funnel_no_outlier$sentence_structure+ma_data_funnel_no_outlier$mean_age, ma_data_funnel_no_outlier$d_var_calc)
f4 <- funnel(ma_model_ss_age, xlab = "effect size", col = ss_colors)
f4_b <- funnel(ma_model_funnel_no_outlier_ss_age, xlab = "effect size", col = ss_colors_no_outlier)
CONTINUOUS_VARS <- c("n_1", "x_1", "sd_1", "d_calc", "d_var_calc", "mean_age","n_train_test_pair","n_test_trial_per_pair","productive_vocab_mean", "productive_vocab_median")
long_continuous <- ma_data %>%
pivot_longer(cols = CONTINUOUS_VARS)
long_continuous %>%
ggplot(aes(x = value)) +
geom_histogram() +
facet_wrap(~ name, scale = "free_x") +
labs(title = "Distribution of continuous measures")
long_continuous %>%
group_by(name) %>%
summarize(mean = mean(value),
sd = sd(value)) %>%
kable()
| name | mean | sd |
|---|---|---|
| d_calc | 0.3329912 | 3.3611587 |
| d_var_calc | 0.4885109 | 1.1410342 |
| mean_age | NA | NA |
| n_1 | 16.9375000 | 5.2341029 |
| n_test_trial_per_pair | 2.3750000 | 0.8062258 |
| n_train_test_pair | 2.8750000 | 1.5000000 |
| productive_vocab_mean | NA | NA |
| productive_vocab_median | NA | NA |
| sd_1 | 0.0793202 | 0.0455952 |
| x_1 | 0.5187500 | 0.1819313 |
CATEGORICAL_VARS <- c("sentence_structure", "agent_argument_type_clean", "patient_argument_type_clean", "stimuli_actor","agent_argument_number","transitive_event_type","intransitive_event_type",
"presentation_type", "character_identification",
"test_mass_or_distributed", "practice_phase", "test_method")
long_categorical <- ma_data %>%
pivot_longer(cols = CATEGORICAL_VARS) %>%
count(name, value) # this is a short cut for group_by() %>% summarize(count = n())
long_categorical %>%
ggplot(aes(x = value, y = n)) +
facet_wrap(~ name, scale = "free_x") +
geom_col(position = 'dodge',width=0.4) +
theme(text = element_text(size=8),
axis.text.x = element_text(angle = 90, hjust = 1)) # rotate x-axis text
ma_data_young <- ma_data_young <- ma_data %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months < 36)
ma_data_old <- ma_data %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months > 36 | age_months == 36)
ma_data_vocab <- ma_data %>%
filter(!is.na(productive_vocab_median))
m1 <- rma.mv(d_calc, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m1)
##
## Multivariate Meta-Analysis Model (k = 16; method: REML)
##
## logLik Deviance AIC BIC AICc
## -70.5838 141.1676 145.1676 146.5837 146.1676
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.2513 1.1186 4 no short_cite
##
## Test for Heterogeneity:
## Q(df = 15) = 218.5173, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## 1.0968 0.5675 1.9326 0.0533 -0.0155 2.2092 .
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_young <- rma.mv(d_calc, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -31.1322 62.2644 66.2644 66.6588 68.2644
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.7519 1.3236 3 no short_cite
##
## Test for Heterogeneity:
## Q(df = 9) = 128.0023, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## 0.9857 0.7726 1.2757 0.2021 -0.5287 2.5000
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_old <- rma.mv(d_calc, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -25.0728 50.1455 54.1455 52.9181 66.1455
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.0000 0.0000 2 no short_cite
##
## Test for Heterogeneity:
## Q(df = 4) = 46.7325, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## 1.4598 0.1861 7.8438 <.0001 1.0950 1.8246 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>%
ggplot(aes(x = mean_age/30.44, y = d_calc,color = unique_id)) +
geom_point() +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months)")
ma_data %>%
ggplot(aes(x = mean_age/30.44, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
geom_smooth(color = "red") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months)") +
theme(legend.position = "none")
m_simple <- rma.mv(d_calc ~ 1, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
m_age <- rma.mv(d_calc ~ mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_simple)
##
## Multivariate Meta-Analysis Model (k = 16; method: REML)
##
## logLik Deviance AIC BIC AICc
## -70.5838 141.1676 145.1676 146.5837 146.1676
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.2513 1.1186 4 no short_cite
##
## Test for Heterogeneity:
## Q(df = 15) = 218.5173, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## 1.0968 0.5675 1.9326 0.0533 -0.0155 2.2092 .
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(m_age)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -58.4528 116.9057 122.9057 124.6005 125.5723
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.2883 1.5127 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 13) = 180.5581, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 4.0575, p-val = 0.0440
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -1.4262 1.4948 -0.9541 0.3400 -4.3559 1.5035
## mean_age 0.0026 0.0013 2.0143 0.0440 0.0001 0.0052 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
forest(m_age,
header = T,
slab = ma_data$unique_id,
col = "red",
cex = .7
)
funnel(m_age)
ma_data_young %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
geom_smooth(color = "red") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months)") +
theme(legend.position = "none")
m_age_young <- rma.mv(d_calc ~ mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -29.4640 58.9279 64.9279 65.1663 70.9279
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.4152 1.1896 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 8) = 110.5672, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.3357, p-val = 0.2478
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.8533 2.5782 1.4946 0.1350 -1.1998 8.9064
## mean_age -0.0037 0.0032 -1.1557 0.2478 -0.0099 0.0026
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
forest(m_age_young,
header = T,
slab = ma_data_young$unique_id,
col = "red",
cex = .7
)
ma_data_old %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
geom_smooth(color = "red") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months)") +
theme(legend.position = "none")
m_age_old <- rma.mv(d_calc ~ mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_age_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -23.1354 46.2708 52.2708 49.5667 76.2708
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 18.3786 4.2870 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 46.1920, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 4.7357, p-val = 0.0295
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -20.5753 10.5482 -1.9506 0.0511 -41.2493 0.0988 .
## mean_age 0.0162 0.0074 2.1762 0.0295 0.0016 0.0307 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
geom_smooth(color = "red") +
ylab("Effect Size") +
xlab("Median Vocabulary Size") +
ggtitle("Syntactical Bootstrapping effect size vs. Median Vocabulary (months)") +
theme(legend.position = "none")
m_age_vocab <- rma.mv(d_calc ~ productive_vocab_median, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_age_vocab)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.6288 11.2575 17.2575 15.4164 41.2575
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.8537 1.3615 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 4) = 51.4381, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1993, p-val = 0.6553
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 1.9083 1.3230 1.4424 0.1492 -0.6848 4.5014
## productive_vocab_median -0.0078 0.0176 -0.4464 0.6553 -0.0423 0.0266
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_vocab_with_age <- rma.mv(d_calc ~ mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_vocab_with_age)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.5064 11.0129 17.0129 15.1718 41.0129
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0957 1.4476 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 4) = 54.7221, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.6046, p-val = 0.4368
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.4753 2.7320 1.2721 0.2033 -1.8793 8.8299
## mean_age -0.0028 0.0035 -0.7775 0.4368 -0.0097 0.0042
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = sentence_structure)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months)")
m_age_sentence <- rma.mv(d_calc ~ mean_age + sentence_structure, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_sentence)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -29.4584 58.9167 66.9167 68.8564 72.6310
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.9134 0.9557 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 97.4627, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 60.8631, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -2.4912 1.1879 -2.0972 0.0360 -4.8193
## mean_age 0.0007 0.0011 0.6583 0.5103 -0.0014
## sentence_structuretransitive 3.1601 0.4150 7.6155 <.0001 2.3468
## ci.ub
## intrcpt -0.1630 *
## mean_age 0.0029
## sentence_structuretransitive 3.9734 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_sentence <- rma.mv(d_calc ~ mean_age * sentence_structure, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_sentence)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -19.9833 39.9666 49.9666 51.9561 61.9666
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.1923 1.0919 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 82.0272, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 76.6859, p-val < .0001
##
## Model Results:
##
## estimate se zval pval
## intrcpt 26.1561 7.2833 3.5912 0.0003
## mean_age -0.0306 0.0079 -3.8607 0.0001
## sentence_structuretransitive -26.2584 7.3523 -3.5714 0.0004
## mean_age:sentence_structuretransitive 0.0321 0.0080 4.0065 <.0001
## ci.lb ci.ub
## intrcpt 11.8811 40.4311 ***
## mean_age -0.0461 -0.0151 ***
## sentence_structuretransitive -40.6686 -11.8481 ***
## mean_age:sentence_structuretransitive 0.0164 0.0478 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, , color = sentence_structure)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), with sentence_structure")
m_age_sentence_young <- rma.mv(d_calc ~ mean_age + sentence_structure, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_sentence_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -8.6221 17.2443 25.2443 25.0279 45.2443
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.1806 1.0865 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 63.3586, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 42.9447, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 1.1530 2.5577 0.4508 0.6521 -3.8600
## mean_age -0.0034 0.0031 -1.0856 0.2777 -0.0095
## sentence_structuretransitive 2.7606 0.4286 6.4408 <.0001 1.9206
## ci.ub
## intrcpt 6.1660
## mean_age 0.0027
## sentence_structuretransitive 3.6007 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_sentence_young <- rma.mv(d_calc ~ mean_age * sentence_structure, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_sentence_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -7.1583 14.3166 24.3166 23.2754 84.3166
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.1752 1.0841 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 60.9361, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 45.2491, p-val < .0001
##
## Model Results:
##
## estimate se zval pval
## intrcpt 88.3469 57.5601 1.5349 0.1248
## mean_age -0.0994 0.0634 -1.5681 0.1168
## sentence_structuretransitive -84.6169 57.6255 -1.4684 0.1420
## mean_age:sentence_structuretransitive 0.0962 0.0635 1.5163 0.1294
## ci.lb ci.ub
## intrcpt -24.4689 201.1626
## mean_age -0.2237 0.0248
## sentence_structuretransitive -197.5608 28.3270
## mean_age:sentence_structuretransitive -0.0282 0.2206
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = sentence_structure)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months)")
m_age_sentence_old <- rma.mv(d_calc ~ mean_age + sentence_structure, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_age_sentence_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -7.3734 14.7468 22.7468 17.5193 62.7468
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 22.2636 4.7184 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 18.1687, p-val = 0.0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 33.6609, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -32.6173 11.0000 -2.9652 0.0030 -54.1770
## mean_age 0.0164 0.0075 2.1768 0.0295 0.0016
## sentence_structuretransitive 11.9559 2.2287 5.3646 <.0001 7.5878
## ci.ub
## intrcpt -11.0577 **
## mean_age 0.0312 *
## sentence_structuretransitive 16.3241 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_sentence_old_interaction <- rma.mv(d_calc ~ mean_age * sentence_structure, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_age_sentence_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -7.3734 14.7468 22.7468 17.5193 62.7468
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 22.2636 4.7184 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 18.1687, p-val = 0.0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 33.6609, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -32.6173 11.0000 -2.9652 0.0030 -54.1770
## mean_age 0.0164 0.0075 2.1768 0.0295 0.0016
## sentence_structuretransitive 11.9559 2.2287 5.3646 <.0001 7.5878
## ci.ub
## intrcpt -11.0577 **
## mean_age 0.0312 *
## sentence_structuretransitive 16.3241 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = sentence_structure)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Median Productive Vocab") +
ggtitle("Syntactical Bootstrapping effect size vs. Median Productive Vocab")
decide to compare noun, pronoun and varying
ma_data %>% group_by(agent_argument_type_clean) %>% count()
## # A tibble: 2 x 2
## # Groups: agent_argument_type_clean [2]
## agent_argument_type_clean n
## <chr> <int>
## 1 noun 4
## 2 pronoun 12
ma_data_at <- ma_data %>% filter(agent_argument_type_clean != "noun_phrase")
ma_data_at_young <- ma_data_at %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months < 36)
ma_data_at_old <- ma_data_at %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months > 36 | age_months == 36)
ma_data_at_vocab <- ma_data_at %>%
filter(!is.na(productive_vocab_median))
ma_data_at %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = agent_argument_type_clean)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by agent argument")
m_age_aa <- rma.mv(d_calc ~ mean_age + agent_argument_type_clean, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_at)
summary(m_age_aa)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -57.6959 115.3919 123.3919 125.3315 129.1062
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.8811 1.6974 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 179.9297, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 4.8924, p-val = 0.0866
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -1.8764 1.6226 -1.1565 0.2475 -5.0566
## mean_age 0.0029 0.0014 2.1208 0.0339 0.0002
## agent_argument_type_cleanpronoun 0.3426 0.4557 0.7518 0.4522 -0.5506
## ci.ub
## intrcpt 1.3037
## mean_age 0.0056 *
## agent_argument_type_cleanpronoun 1.2358
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_aa_interaction <- rma.mv(d_calc ~ mean_age * agent_argument_type_clean, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_at)
summary(m_age_aa_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -53.0513 106.1027 116.1027 118.0922 128.1027
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 4.5123 2.1242 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 176.8145, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 14.1777, p-val = 0.0027
##
## Model Results:
##
## estimate se zval pval
## intrcpt 10.8558 4.8120 2.2560 0.0241
## mean_age -0.0149 0.0063 -2.3531 0.0186
## agent_argument_type_cleanpronoun -14.0663 4.9322 -2.8520 0.0043
## mean_age:agent_argument_type_cleanpronoun 0.0193 0.0066 2.9445 0.0032
## ci.lb ci.ub
## intrcpt 1.4245 20.2872 *
## mean_age -0.0274 -0.0025 *
## agent_argument_type_cleanpronoun -23.7331 -4.3994 **
## mean_age:agent_argument_type_cleanpronoun 0.0065 0.0322 **
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_at_young %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = agent_argument_type_clean)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by agent argument, young only")
m_age_aa_young <- rma.mv(d_calc ~ mean_age + agent_argument_type_clean, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_at_young)
summary(m_age_aa_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -29.0641 58.1282 66.1282 65.9118 86.1282
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.5428 1.2421 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 101.5135, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 1.3567, p-val = 0.5074
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 3.5883 2.7415 1.3089 0.1906 -1.7849
## mean_age -0.0034 0.0033 -1.0332 0.3015 -0.0099
## agent_argument_type_cleanpronoun 0.1255 0.4653 0.2696 0.7874 -0.7865
## ci.ub
## intrcpt 8.9616
## mean_age 0.0031
## agent_argument_type_cleanpronoun 1.0374
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_aa_young_interaction <- rma.mv(d_calc ~ mean_age * agent_argument_type_clean, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_at_young)
summary(m_age_aa_young_interaction)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -26.3676 52.7351 62.7351 61.6939 122.7351
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 3.2854 1.8126 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 85.7429, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 5.8364, p-val = 0.1198
##
## Model Results:
##
## estimate se zval pval
## intrcpt 11.2717 4.7655 2.3653 0.0180
## mean_age -0.0144 0.0063 -2.2815 0.0225
## agent_argument_type_cleanpronoun -12.3929 5.7651 -2.1497 0.0316
## mean_age:agent_argument_type_cleanpronoun 0.0170 0.0078 2.1826 0.0291
## ci.lb ci.ub
## intrcpt 1.9315 20.6119 *
## mean_age -0.0269 -0.0020 *
## agent_argument_type_cleanpronoun -23.6923 -1.0936 *
## mean_age:agent_argument_type_cleanpronoun 0.0017 0.0323 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_at_old %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = agent_argument_type_clean)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by agent argument, young only")
{r simple_model_age_aat_old} # m_age_aa_old <- rma.mv(d_calc ~ mean_age + agent_argument_type_clean, V = d_var_calc, # random = ~ 1 | short_cite, data = ma_data_at_old) # summary(m_age_aa_old) #{r interaction_model_age_old} # m_age_aa_old_interaction <- rma.mv(d_calc ~ mean_age * agent_argument_type_clean, V = d_var_calc, # random = ~ 1 | short_cite, data = ma_data_at_old) # summary(m_age_aa_old_interaction) #ma_data_at_vocab %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = agent_argument_type_clean)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by agent argument, with vocab")
m_age_aa_vocab <- rma.mv(d_calc ~ productive_vocab_median + agent_argument_type_clean, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_at_vocab)
summary(m_age_aa_vocab)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.0814 10.1628 18.1628 14.5573 58.1628
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.6062 1.2673 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 32.8805, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 0.6963, p-val = 0.7060
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 1.6986 1.3116 1.2951 0.1953 -0.8720
## productive_vocab_median -0.0053 0.0180 -0.2933 0.7693 -0.0405
## agent_argument_type_cleanpronoun 0.3322 0.4742 0.7006 0.4835 -0.5971
## ci.ub
## intrcpt 4.2692
## productive_vocab_median 0.0300
## agent_argument_type_cleanpronoun 1.2615
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_aa_vocab_interaction <- rma.mv(d_calc ~ productive_vocab_median * agent_argument_type_clean, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_at_vocab)
summary(m_age_aa_vocab_interaction)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -1.4686 2.9373 12.9373 6.4030 72.9373
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.2783 1.1306 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 20.6776, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 7.2171, p-val = 0.0653
##
## Model Results:
##
## estimate se
## intrcpt 5.9771 2.0944
## productive_vocab_median -0.0870 0.0368
## agent_argument_type_cleanpronoun -5.0090 2.1579
## productive_vocab_median:agent_argument_type_cleanpronoun 0.1072 0.0422
## zval pval
## intrcpt 2.8539 0.0043
## productive_vocab_median -2.3653 0.0180
## agent_argument_type_cleanpronoun -2.3212 0.0203
## productive_vocab_median:agent_argument_type_cleanpronoun 2.5432 0.0110
## ci.lb ci.ub
## intrcpt 1.8722 10.0820 **
## productive_vocab_median -0.1592 -0.0149 *
## agent_argument_type_cleanpronoun -9.2384 -0.7796 *
## productive_vocab_median:agent_argument_type_cleanpronoun 0.0246 0.1899 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>% group_by(patient_argument_type_clean) %>% count()
## # A tibble: 4 x 2
## # Groups: patient_argument_type_clean [4]
## patient_argument_type_clean n
## <chr> <int>
## 1 intransitive 4
## 2 noun 5
## 3 noun_phrase 3
## 4 pronoun 4
ma_data %>% group_by(agent_argument_number) %>% count()
## # A tibble: 1 x 2
## # Groups: agent_argument_number [1]
## agent_argument_number n
## <chr> <int>
## 1 1 16
ma_data %>% ggplot(aes(x = n_repetitions_sentence)) +
geom_histogram()
### all
ma_data %>%
ggplot(aes(x = n_repetitions_sentence, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("number of reptition in sentence") +
ggtitle("Syntactical Bootstrapping effect size vs. number of sentence repetition")
m_rep <- rma.mv(d_calc ~ n_repetitions_sentence, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_rep)
##
## Multivariate Meta-Analysis Model (k = 16; method: REML)
##
## logLik Deviance AIC BIC AICc
## -67.4324 134.8648 140.8648 142.7820 143.2648
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.5197 0.7209 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 14) = 147.4682, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 4.8951, p-val = 0.0269
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.7818 1.2742 2.9680 0.0030 1.2845 6.2792 **
## n_repetitions_sentence -0.4294 0.1941 -2.2125 0.0269 -0.8098 -0.0490 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_age <- rma.mv(d_calc ~ n_repetitions_sentence + mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_rep_age)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -56.2046 112.4091 120.4091 122.3488 126.1234
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 3.5811 1.8924 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 128.1286, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 4.6401, p-val = 0.0983
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -0.7678 3.8922 -0.1973 0.8436 -8.3964 6.8607
## n_repetitions_sentence -0.1440 0.5095 -0.2826 0.7775 -1.1426 0.8547
## mean_age 0.0029 0.0014 2.0098 0.0445 0.0001 0.0057 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_age_interaction <- rma.mv(d_calc ~ n_repetitions_sentence * mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_rep_age_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -51.8655 103.7311 113.7311 115.7206 125.7311
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.0000 0.0000 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 100.5837, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 84.2023, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 13.2170 1.8064 7.3166 <.0001 9.6765
## n_repetitions_sentence -2.0554 0.3165 -6.4934 <.0001 -2.6758
## mean_age -0.0086 0.0016 -5.5146 <.0001 -0.0117
## n_repetitions_sentence:mean_age 0.0016 0.0003 5.2483 <.0001 0.0010
## ci.ub
## intrcpt 16.7576 ***
## n_repetitions_sentence -1.4350 ***
## mean_age -0.0056 ***
## n_repetitions_sentence:mean_age 0.0022 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
ggplot(aes(x = n_repetitions_sentence, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("number of reptition in sentence") +
ggtitle("Syntactical Bootstrapping effect size vs. number of sentence repetition, young only")
m_rep_young <- rma.mv(d_calc ~ n_repetitions_sentence, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_rep_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -27.8726 55.7452 61.7452 61.9835 67.7452
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.2950 0.5431 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 8) = 62.8092, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 9.5503, p-val = 0.0020
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 5.4513 1.4904 3.6576 0.0003 2.5301 8.3724
## n_repetitions_sentence -0.6387 0.2067 -3.0904 0.0020 -1.0438 -0.2336
##
## intrcpt ***
## n_repetitions_sentence **
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_age_young <- rma.mv(d_calc ~ n_repetitions_sentence + mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_rep_age_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -27.3902 54.7805 62.7805 62.5641 82.7805
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.6001 0.7746 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 56.6214, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.3060, p-val = 0.0704
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 6.1886 2.5236 2.4524 0.0142 1.2426 11.1347 *
## n_repetitions_sentence -0.5671 0.3209 -1.7676 0.0771 -1.1960 0.0617 .
## mean_age -0.0016 0.0032 -0.4902 0.6240 -0.0079 0.0048
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_age_interaction_young <- rma.mv(d_calc ~ n_repetitions_sentence * mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_rep_age_interaction_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -26.0991 52.1981 62.1981 61.1569 122.1981
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.0000 0.0003 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 51.8353, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 76.1669, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 18.5392 6.9010 2.6865 0.0072 5.0136
## n_repetitions_sentence -2.7701 0.9006 -3.0759 0.0021 -4.5353
## mean_age -0.0155 0.0090 -1.7117 0.0870 -0.0331
## n_repetitions_sentence:mean_age 0.0025 0.0011 2.1877 0.0287 0.0003
## ci.ub
## intrcpt 32.0648 **
## n_repetitions_sentence -1.0050 **
## mean_age 0.0022 .
## n_repetitions_sentence:mean_age 0.0047 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
ggplot(aes(x = n_repetitions_sentence, y = d_calc, size = n_1)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("number of reptition in sentence") +
ggtitle("Syntactical Bootstrapping effect size vs. number of sentence repetition, old only")
m_rep_old <- rma.mv(d_calc ~ n_repetitions_sentence, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_rep_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -24.6833 49.3666 55.3666 52.6625 79.3666
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.6349 3.1040 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 46.6287, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0014, p-val = 0.9703
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 1.6173 6.1366 0.2636 0.7921 -10.4101 13.6447
## n_repetitions_sentence -0.0329 0.8839 -0.0372 0.9703 -1.7652 1.6995
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_age_old <- rma.mv(d_calc ~ n_repetitions_sentence + mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_rep_age_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -21.7925 43.5851 51.5851 46.3577 91.5851
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.4487 3.0739 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 40.8937, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.7364, p-val = 0.0568
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -34.0896 16.1013 -2.1172 0.0342 -65.6476 -2.5316
## n_repetitions_sentence 1.3462 1.0478 1.2848 0.1989 -0.7075 3.4000
## mean_age 0.0197 0.0082 2.3948 0.0166 0.0036 0.0358
##
## intrcpt *
## n_repetitions_sentence
## mean_age *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_age_interaction_old <- rma.mv(d_calc ~ n_repetitions_sentence * mean_age, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_rep_age_interaction_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -21.7925 43.5851 51.5851 46.3577 91.5851
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.4487 3.0739 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 40.8937, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.7364, p-val = 0.0568
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -34.0896 16.1013 -2.1172 0.0342 -65.6476 -2.5316
## n_repetitions_sentence 1.3462 1.0478 1.2848 0.1989 -0.7075 3.4000
## mean_age 0.0197 0.0082 2.3948 0.0166 0.0036 0.0358
##
## intrcpt *
## n_repetitions_sentence
## mean_age *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_v_r <- rma.mv(d_calc ~ n_repetitions_sentence + productive_vocab_median, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_rep_v_r)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -3.4583 6.9166 14.9166 11.3111 54.9166
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.1235 0.3514 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 7.0746, p-val = 0.0696
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 12.2593, p-val = 0.0022
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 7.7187 1.8712 4.1250 <.0001 4.0512 11.3862
## n_repetitions_sentence -0.9738 0.2883 -3.3780 0.0007 -1.5388 -0.4088
## productive_vocab_median -0.0071 0.0176 -0.4014 0.6882 -0.0415 0.0274
##
## intrcpt ***
## n_repetitions_sentence ***
## productive_vocab_median
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_rep_v_r_interaction <- rma.mv(d_calc ~ n_repetitions_sentence *productive_vocab_median, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_rep_v_r)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -3.4583 6.9166 14.9166 11.3111 54.9166
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.1235 0.3514 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 7.0746, p-val = 0.0696
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 12.2593, p-val = 0.0022
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 7.7187 1.8712 4.1250 <.0001 4.0512 11.3862
## n_repetitions_sentence -0.9738 0.2883 -3.3780 0.0007 -1.5388 -0.4088
## productive_vocab_median -0.0071 0.0176 -0.4014 0.6882 -0.0415 0.0274
##
## intrcpt ***
## n_repetitions_sentence ***
## productive_vocab_median
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_modality)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli modality")
m_stim_mod<- rma.mv(d_calc ~ mean_age + stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_stim_mod)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -56.5931 113.1862 121.1862 123.1258 126.9005
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 3.9820 1.9955 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 172.7654, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 4.6822, p-val = 0.0962
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -1.6445 2.2452 -0.7325 0.4639 -6.0450 2.7560
## mean_age 0.0031 0.0015 2.1320 0.0330 0.0003 0.0060 *
## stimuli_modalityvideo -0.3061 2.3757 -0.1288 0.8975 -4.9623 4.3502
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_mod_interaction<- rma.mv(d_calc ~ mean_age * stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_stim_mod_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -57.1955 114.3910 124.3910 126.3805 136.3910
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 3.9761 1.9940 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 172.7393, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 4.6845, p-val = 0.1964
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -2.9504 21.4977 -0.1372 0.8908 -45.0852
## mean_age 0.0050 0.0307 0.1624 0.8710 -0.0551
## stimuli_modalityvideo 1.0054 21.5856 0.0466 0.9628 -41.3015
## mean_age:stimuli_modalityvideo -0.0019 0.0307 -0.0611 0.9513 -0.0620
## ci.ub
## intrcpt 39.1844
## mean_age 0.0651
## stimuli_modalityvideo 43.3124
## mean_age:stimuli_modalityvideo 0.0583
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_modality)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli modality, young")
m_stim_mod_young<- rma.mv(d_calc ~ mean_age + stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_stim_mod_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -27.6415 55.2829 63.2829 63.0666 83.2829
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.7770 1.3330 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 59.6530, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 1.7056, p-val = 0.4262
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.4120 2.6939 1.2666 0.2053 -1.8680 8.6921
## mean_age -0.0041 0.0033 -1.2360 0.2164 -0.0107 0.0024
## stimuli_modalityvideo 1.1958 1.6962 0.7050 0.4808 -2.1287 4.5203
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_mod_interaction_young <- rma.mv(d_calc ~ mean_age * stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_stim_mod_interaction_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.1963 56.3926 66.3926 65.3514 126.3926
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.7348 1.3171 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 59.3453, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 1.8212, p-val = 0.6103
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -2.9504 21.4456 -0.1376 0.8906 -44.9829
## mean_age 0.0050 0.0307 0.1624 0.8710 -0.0551
## stimuli_modalityvideo 7.6679 21.6423 0.3543 0.7231 -34.7503
## mean_age:stimuli_modalityvideo -0.0092 0.0308 -0.2998 0.7643 -0.0697
## ci.ub
## intrcpt 39.0821
## mean_age 0.0651
## stimuli_modalityvideo 50.0860
## mean_age:stimuli_modalityvideo 0.0512
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_modality)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli modality, old")
ma_data_vocab %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = stimuli_modality)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("median vocab") +
ggtitle("Syntactical Bootstrapping effect size vs. median vocab, breakdown by stimuli modality, vocab")
m_stim_mod_v <- rma.mv(d_calc ~ productive_vocab_median + stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_mod_v)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -3.4583 6.9166 14.9166 11.3111 54.9166
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.1235 0.3514 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 7.0746, p-val = 0.0696
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 12.2593, p-val = 0.0022
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 0.9023 1.0159 0.8881 0.3745 -1.0889 2.8935
## productive_vocab_median -0.0071 0.0176 -0.4014 0.6882 -0.0415 0.0274
## stimuli_modalityvideo 1.9476 0.5765 3.3780 0.0007 0.8175 3.0776
##
## intrcpt
## productive_vocab_median
## stimuli_modalityvideo ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_mod_intereaction_v <- rma.mv(d_calc ~ productive_vocab_median * stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_mod_intereaction_v)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -4.0472 8.0945 18.0945 11.5602 78.0945
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.1237 0.3517 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 7.0656, p-val = 0.0292
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 12.2511, p-val = 0.0066
##
## Model Results:
##
## estimate se zval
## intrcpt 1.4167 5.5073 0.2572
## productive_vocab_median -0.0167 0.1027 -0.1624
## stimuli_modalityvideo 1.4191 5.5910 0.2538
## productive_vocab_median:stimuli_modalityvideo 0.0099 0.1042 0.0950
## pval ci.lb ci.ub
## intrcpt 0.7970 -9.3774 12.2107
## productive_vocab_median 0.8710 -0.2179 0.1845
## stimuli_modalityvideo 0.7996 -9.5390 12.3771
## productive_vocab_median:stimuli_modalityvideo 0.9243 -0.1943 0.2141
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_modality)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli modality, age compare vocab")
m_stim_mod_v_a<- rma.mv(d_calc ~ mean_age + stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_mod_v_a)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -3.2187 6.4373 14.4373 10.8318 54.4373
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.1094 0.3308 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 6.5517, p-val = 0.0876
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 13.9246, p-val = 0.0009
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 2.5762 2.5072 1.0275 0.3042 -2.3378 7.4901
## mean_age -0.0029 0.0036 -0.8270 0.4082 -0.0099 0.0040
## stimuli_modalityvideo 2.0689 0.5547 3.7294 0.0002 0.9816 3.1561 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_mod_interaction_v_a <- rma.mv(d_calc ~ mean_age * stimuli_modality, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_mod_interaction_v_a)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -3.7849 7.5698 17.5698 11.0356 77.5698
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.1088 0.3299 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 6.4842, p-val = 0.0391
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 14.0474, p-val = 0.0028
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -2.9504 21.4076 -0.1378 0.8904 -44.9086
## mean_age 0.0050 0.0307 0.1624 0.8710 -0.0551
## stimuli_modalityvideo 7.6739 21.5690 0.3558 0.7220 -34.6006
## mean_age:stimuli_modalityvideo -0.0080 0.0309 -0.2599 0.7949 -0.0685
## ci.ub
## intrcpt 39.0077
## mean_age 0.0651
## stimuli_modalityvideo 49.9483
## mean_age:stimuli_modalityvideo 0.0525
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_actor)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli actor, all")
m_stim_actor_age <- rma.mv(d_calc ~ mean_age + stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_stim_actor_age)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -55.8743 111.7485 119.7485 121.6882 125.4628
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 4.5635 2.1362 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 175.1689, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 8.5454, p-val = 0.0139
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -3.4879 1.9468 -1.7916 0.0732 -7.3036 0.3278 .
## mean_age 0.0044 0.0016 2.7916 0.0052 0.0013 0.0075 **
## stimuli_actorperson 1.0343 0.5339 1.9372 0.0527 -0.0121 2.0808 .
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_actor_age_interaction <- rma.mv(d_calc ~ mean_age * stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_stim_actor_age_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -54.3997 108.7994 118.7994 120.7889 130.7994
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 5.7049 2.3885 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 161.4404, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 9.0029, p-val = 0.0293
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -3.1994 3.2878 -0.9731 0.3305 -9.6434
## mean_age 0.0041 0.0032 1.2655 0.2057 -0.0022
## stimuli_actorperson 0.5305 2.7608 0.1921 0.8476 -4.8807
## mean_age:stimuli_actorperson 0.0007 0.0036 0.2003 0.8413 -0.0063
## ci.ub
## intrcpt 3.2446
## mean_age 0.0104
## stimuli_actorperson 5.9416
## mean_age:stimuli_actorperson 0.0078
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_actor)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli actor, young")
m_stim_actor_age_young <- rma.mv(d_calc ~ mean_age + stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_stim_actor_age_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.4463 56.8926 64.8926 64.6763 84.8926
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.3823 1.5435 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 108.8621, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 1.0896, p-val = 0.5800
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 4.0058 4.9490 0.8094 0.4183 -5.6941 13.7056
## mean_age -0.0038 0.0058 -0.6565 0.5115 -0.0153 0.0076
## stimuli_actorperson -0.0689 0.8519 -0.0809 0.9355 -1.7386 1.6008
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_actor_age_interaction_young <- rma.mv(d_calc ~ mean_age * stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_stim_actor_age_interaction_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -26.8142 53.6283 63.6283 62.5871 123.6283
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.3555 1.1643 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 53.3803, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 2.2000, p-val = 0.5320
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -6.8631 13.1352 -0.5225 0.6013 -32.6076
## mean_age 0.0111 0.0178 0.6247 0.5322 -0.0238
## stimuli_actorperson 13.8307 15.5787 0.8878 0.3747 -16.7031
## mean_age:stimuli_actorperson -0.0186 0.0208 -0.8966 0.3699 -0.0593
## ci.ub
## intrcpt 18.8815
## mean_age 0.0461
## stimuli_actorperson 44.3644
## mean_age:stimuli_actorperson 0.0221
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_actor)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli actor, old")
m_stim_actor_age_old <- rma.mv(d_calc ~ mean_age + stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_stim_actor_age_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -21.7925 43.5851 51.5851 46.3577 91.5851
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.4487 3.0739 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 40.8937, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.7364, p-val = 0.0568
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -28.7046 12.9777 -2.2118 0.0270 -54.1404 -3.2688 *
## mean_age 0.0197 0.0082 2.3948 0.0166 0.0036 0.0358 *
## stimuli_actorperson 6.7312 5.2392 1.2848 0.1989 -3.5374 16.9998
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_actor_age_interaction_old <- rma.mv(d_calc ~ mean_age * stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_stim_actor_age_interaction_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -21.7925 43.5851 51.5851 46.3577 91.5851
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.4487 3.0739 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 40.8937, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.7364, p-val = 0.0568
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -28.7046 12.9777 -2.2118 0.0270 -54.1404 -3.2688 *
## mean_age 0.0197 0.0082 2.3948 0.0166 0.0036 0.0358 *
## stimuli_actorperson 6.7312 5.2392 1.2848 0.1989 -3.5374 16.9998
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = stimuli_actor)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli actor, vocab")
m_stim_actor_vocab <- rma.mv(d_calc ~ productive_vocab_median + stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_actor_vocab)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -3.0526 6.1052 14.1052 10.4997 54.1052
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.8983 0.9478 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 15.2192, p-val = 0.0016
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.0929, p-val = 0.0784
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -5.1831 3.4257 -1.5130 0.1303 -11.8973 1.5310
## productive_vocab_median 0.1190 0.0605 1.9652 0.0494 0.0003 0.2376 *
## stimuli_actorperson 3.7386 1.6975 2.2024 0.0276 0.4115 7.0657 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_actor_vocab_interaction <- rma.mv(d_calc ~ productive_vocab_median * stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_actor_vocab_interaction)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -1.8145 3.6289 13.6289 7.0947 73.6289
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.0695 1.0342 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 14.8974, p-val = 0.0006
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 6.5762, p-val = 0.0867
##
## Model Results:
##
## estimate se zval pval
## intrcpt -3.5249 3.6548 -0.9645 0.3348
## productive_vocab_median 0.0891 0.0645 1.3811 0.1673
## stimuli_actorperson -4.5120 6.5909 -0.6846 0.4936
## productive_vocab_median:stimuli_actorperson 0.2442 0.1897 1.2877 0.1978
## ci.lb ci.ub
## intrcpt -10.6882 3.6384
## productive_vocab_median -0.0374 0.2156
## stimuli_actorperson -17.4299 8.4058
## productive_vocab_median:stimuli_actorperson -0.1275 0.6159
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = stimuli_actor)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by stimuli actor, age vocab comparison")
m_stim_actor_vocab_a <- rma.mv(d_calc ~ mean_age + stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_actor_vocab_a)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -2.8031 5.6062 13.6062 10.0007 53.6062
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.0000 0.0000 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 5.7871, p-val = 0.1224
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 51.8748, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -16.4159 3.0466 -5.3882 <.0001 -22.3872 -10.4446 ***
## mean_age 0.0243 0.0043 5.7021 <.0001 0.0159 0.0326 ***
## stimuli_actorperson 3.8266 0.5470 6.9954 <.0001 2.7545 4.8988 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_stim_actor_vocab_interaction_a <- rma.mv(d_calc ~ mean_age * stimuli_actor, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_stim_actor_vocab_interaction_a)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -1.4159 2.8317 12.8317 6.2974 72.8317
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 0.3194 0.5651 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 3.9127, p-val = 0.1414
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 8.9325, p-val = 0.0302
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -13.4712 7.8773 -1.7101 0.0872 -28.9105
## mean_age 0.0203 0.0107 1.8867 0.0592 -0.0008
## stimuli_actorperson -40.8319 31.5812 -1.2929 0.1960 -102.7298
## mean_age:stimuli_actorperson 0.0695 0.0492 1.4124 0.1578 -0.0269
## ci.ub
## intrcpt 1.9680 .
## mean_age 0.0413 .
## stimuli_actorperson 21.0660
## mean_age:stimuli_actorperson 0.1659
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>% count(transitive_event_type)
## # A tibble: 1 x 2
## transitive_event_type n
## <chr> <int>
## 1 AGENT 16
m_data_transtivity <- ma_data %>%
filter(transitive_event_type != "minimal_contact")
m_data_transtivity_young <- ma_data %>%
filter(transitive_event_type != "minimal_contact") %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months < 36)
m_data_transtivity_old <- ma_data %>%
filter(transitive_event_type != "minimal_contact") %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months > 36 | age_months == 36)
m_data_transitivity_vocab <- ma_data_vocab %>%
filter(transitive_event_type != "minimal_contact")
m_data_transtivity %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = transitive_event_type)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by transitive_event_type")
ma_data %>% count(intransitive_event_type)
## # A tibble: 1 x 2
## intransitive_event_type n
## <chr> <int>
## 1 AGENT 16
m_data_intran <- ma_data %>%
filter(intransitive_event_type != "minimal_contact")
m_data_intran_young <- ma_data %>%
filter(intransitive_event_type != "minimal_contact") %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months < 36)
m_data_intran_old <- ma_data %>%
filter(intransitive_event_type != "minimal_contact") %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months > 36 | age_months == 36)
m_data_intran_vocab <- ma_data_vocab %>%
filter(intransitive_event_type != "minimal_contact")
m_data_intran %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = intransitive_event_type)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by intransitive_event_type")
ma_data %>% group_by(transitive_event_type, intransitive_event_type) %>% count()
## # A tibble: 1 x 3
## # Groups: transitive_event_type, intransitive_event_type [1]
## transitive_event_type intransitive_event_type n
## <chr> <chr> <int>
## 1 AGENT AGENT 16
ma_data %>% ggplot(aes(x = n_repetitions_video)) +
geom_histogram()
ma_data %>% count(test_method)
## # A tibble: 2 x 2
## test_method n
## <chr> <int>
## 1 look 6
## 2 point 10
ma_data %>% count(presentation_type)
## # A tibble: 1 x 2
## presentation_type n
## <chr> <int>
## 1 immediate_after 16
ma_data_pt <- ma_data %>%
filter(presentation_type != "simultaneous")
ma_data_pt_young <- ma_data_pt %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months < 36)
ma_data_pt_old <- ma_data_pt %>%
mutate(age_months = mean_age/30.44) %>%
filter(age_months > 36 | age_months == 36)
ma_data_pt_vocab <- ma_data_vocab %>%
filter(presentation_type != "simultaneous")
ma_data_pt %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = presentation_type)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by presentation type")
ma_data %>% count(character_identification)
## # A tibble: 2 x 2
## character_identification n
## <chr> <int>
## 1 no 7
## 2 yes 9
ma_data %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = character_identification)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by character_identification")
m_age_ci <- rma.mv(d_calc ~ mean_age + character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_ci)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -56.0340 112.0680 120.0680 122.0076 125.7823
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 3.2120 1.7922 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 154.5945, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 4.8078, p-val = 0.0904
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -2.6048 2.2311 -1.1675 0.2430 -6.9777 1.7681
## mean_age 0.0029 0.0014 2.0943 0.0362 0.0002 0.0056
## character_identificationyes 1.2510 2.0812 0.6011 0.5478 -2.8282 5.3302
##
## intrcpt
## mean_age *
## character_identificationyes
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_stimuli_ci_interaction <- rma.mv(d_calc ~ mean_age * character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_stimuli_ci_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -53.6799 107.3597 117.3597 119.3492 129.3597
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.5416 1.2416 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 148.8517, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 8.1275, p-val = 0.0434
##
## Model Results:
##
## estimate se zval pval
## intrcpt -4.6041 2.1667 -2.1249 0.0336
## mean_age 0.0050 0.0019 2.6927 0.0071
## character_identificationyes 5.9212 2.7908 2.1217 0.0339
## mean_age:character_identificationyes -0.0048 0.0025 -1.9509 0.0511
## ci.lb ci.ub
## intrcpt -8.8508 -0.3574 *
## mean_age 0.0014 0.0086 **
## character_identificationyes 0.4514 11.3911 *
## mean_age:character_identificationyes -0.0096 0.0000 .
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = character_identification)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by character_identification, young only")
m_age_ci_young <- rma.mv(d_calc ~ mean_age + character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_ci_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.3250 56.6500 64.6500 64.4337 84.6500
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0997 1.4490 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 100.0259, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 1.4077, p-val = 0.4947
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 2.5154 3.5290 0.7128 0.4760 -4.4013 9.4321
## mean_age -0.0028 0.0035 -0.8014 0.4229 -0.0097 0.0041
## character_identificationyes 1.0084 1.9222 0.5246 0.5999 -2.7590 4.7757
##
## intrcpt
## mean_age
## character_identificationyes
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_ci_interaction_young <- rma.mv(d_calc ~ mean_age * character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_ci_interaction_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.3118 56.6236 66.6236 65.5824 126.6236
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0957 1.4476 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 99.8371, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 1.4401, p-val = 0.6962
##
## Model Results:
##
## estimate se zval pval
## intrcpt 7.7218 29.7396 0.2596 0.7951
## mean_age -0.0085 0.0326 -0.2619 0.7934
## character_identificationyes -4.2465 29.8649 -0.1422 0.8869
## mean_age:character_identificationyes 0.0058 0.0328 0.1763 0.8600
## ci.lb ci.ub
## intrcpt -50.5668 66.0105
## mean_age -0.0725 0.0554
## character_identificationyes -62.7806 54.2875
## mean_age:character_identificationyes -0.0585 0.0701
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = character_identification)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by character_identification, old only")
m_age_ci_old <- rma.mv(d_calc ~ mean_age + character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_age_ci_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -21.7925 43.5851 51.5851 46.3577 91.5851
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.4487 3.0739 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 40.8937, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.7364, p-val = 0.0568
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -21.9734 10.2122 -2.1517 0.0314 -41.9890
## mean_age 0.0197 0.0082 2.3948 0.0166 0.0036
## character_identificationyes -6.7312 5.2392 -1.2848 0.1989 -16.9998
## ci.ub
## intrcpt -1.9579 *
## mean_age 0.0358 *
## character_identificationyes 3.5375
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_ci_interaction_old <- rma.mv(d_calc ~ mean_age * character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_old)
summary(m_age_ci_interaction_old)
##
## Multivariate Meta-Analysis Model (k = 5; method: REML)
##
## logLik Deviance AIC BIC AICc
## -21.7925 43.5851 51.5851 46.3577 91.5851
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 9.4487 3.0739 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 2) = 40.8937, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 5.7364, p-val = 0.0568
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -21.9734 10.2122 -2.1517 0.0314 -41.9890
## mean_age 0.0197 0.0082 2.3948 0.0166 0.0036
## character_identificationyes -6.7312 5.2392 -1.2848 0.1989 -16.9998
## ci.ub
## intrcpt -1.9579 *
## mean_age 0.0358 *
## character_identificationyes 3.5375
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = character_identification)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("median vocab") +
ggtitle("Syntactical Bootstrapping effect size vs. median voca, breakdown by character_identification, old only")
ma_data %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = practice_phase)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by practice_phase")
m_age_pf <- rma.mv(d_calc ~ mean_age + character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_ci)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -56.0340 112.0680 120.0680 122.0076 125.7823
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 3.2120 1.7922 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 154.5945, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 4.8078, p-val = 0.0904
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -2.6048 2.2311 -1.1675 0.2430 -6.9777 1.7681
## mean_age 0.0029 0.0014 2.0943 0.0362 0.0002 0.0056
## character_identificationyes 1.2510 2.0812 0.6011 0.5478 -2.8282 5.3302
##
## intrcpt
## mean_age *
## character_identificationyes
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_pf_interaction <- rma.mv(d_calc ~ mean_age * character_identification, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_stimuli_ci_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -53.6799 107.3597 117.3597 119.3492 129.3597
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.5416 1.2416 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 148.8517, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 8.1275, p-val = 0.0434
##
## Model Results:
##
## estimate se zval pval
## intrcpt -4.6041 2.1667 -2.1249 0.0336
## mean_age 0.0050 0.0019 2.6927 0.0071
## character_identificationyes 5.9212 2.7908 2.1217 0.0339
## mean_age:character_identificationyes -0.0048 0.0025 -1.9509 0.0511
## ci.lb ci.ub
## intrcpt -8.8508 -0.3574 *
## mean_age 0.0014 0.0086 **
## character_identificationyes 0.4514 11.3911 *
## mean_age:character_identificationyes -0.0096 0.0000 .
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = practice_phase)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by practice_phase, young only")
m_age_pf_young <- rma.mv(d_calc ~ mean_age + practice_phase, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_pf_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -29.5512 59.1025 67.1025 66.8861 87.1025
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.2976 1.1391 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 91.2605, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 1.6345, p-val = 0.4416
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.8596 2.5519 1.5124 0.1304 -1.1420 8.8613
## mean_age -0.0038 0.0032 -1.1972 0.2312 -0.0100 0.0024
## practice_phaseyes 0.1490 0.3029 0.4920 0.6227 -0.4446 0.7426
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_pf_interaction_young <- rma.mv(d_calc ~ mean_age * practice_phase, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_pf_interaction_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.6312 57.2625 67.2625 66.2213 127.2625
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.9011 1.3788 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 89.2202, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 1.6196, p-val = 0.6550
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 6.4798 5.9731 1.0848 0.2780 -5.2273
## mean_age -0.0072 0.0077 -0.9340 0.3503 -0.0223
## practice_phaseyes -2.8830 5.8906 -0.4894 0.6245 -14.4284
## mean_age:practice_phaseyes 0.0043 0.0084 0.5121 0.6086 -0.0122
## ci.ub
## intrcpt 18.1869
## mean_age 0.0079
## practice_phaseyes 8.6623
## mean_age:practice_phaseyes 0.0208
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = practice_phase)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by practice_phase, old only")
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = practice_phase)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("productive vocab") +
ggtitle("Syntactical Bootstrapping effect size vs. productive vocab, breakdown by practice_phase, old only")
m_age_pf_vocab <- rma.mv(d_calc ~ productive_vocab_median + practice_phase, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_age_pf_vocab)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.9892 11.9784 19.9784 16.3729 59.9784
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.8171 1.3480 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 42.3702, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 0.2288, p-val = 0.8919
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 1.8427 1.3729 1.3422 0.1795 -0.8481 4.5335
## productive_vocab_median -0.0073 0.0178 -0.4110 0.6811 -0.0423 0.0276
## practice_phaseyes 0.0528 0.3117 0.1694 0.8655 -0.5582 0.6638
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_pf_interaction_vocab <- rma.mv(d_calc ~ productive_vocab_median * practice_phase, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_age_pf_interaction_vocab)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.9892 11.9784 19.9784 16.3729 59.9784
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.8171 1.3480 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 42.3702, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 0.2288, p-val = 0.8919
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 1.8427 1.3729 1.3422 0.1795 -0.8481 4.5335
## productive_vocab_median -0.0073 0.0178 -0.4110 0.6811 -0.0423 0.0276
## practice_phaseyes 0.0528 0.3117 0.1694 0.8655 -0.5582 0.6638
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = mean_age, y = d_calc, size = n_1, color = practice_phase)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("mean age") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by practice_phase, old only")
m_age_pf_vocab_a <- rma.mv(d_calc ~ mean_age + practice_phase, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_age_pf_vocab_a)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.8008 11.6016 19.6016 15.9961 59.6016
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0081 1.4171 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 44.0717, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 0.7115, p-val = 0.7007
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.4940 2.7248 1.2823 0.1997 -1.8466 8.8345
## mean_age -0.0029 0.0036 -0.8095 0.4182 -0.0099 0.0041
## practice_phaseyes 0.1027 0.3093 0.3320 0.7399 -0.5035 0.7088
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_pf_interaction_vocab_a <- rma.mv(d_calc ~ mean_age * practice_phase, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_vocab)
summary(m_age_pf_interaction_vocab_a)
##
## Multivariate Meta-Analysis Model (k = 6; method: REML)
##
## logLik Deviance AIC BIC AICc
## -5.8008 11.6016 19.6016 15.9961 59.6016
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0081 1.4171 2 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 44.0717, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 0.7115, p-val = 0.7007
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 3.4940 2.7248 1.2823 0.1997 -1.8466 8.8345
## mean_age -0.0029 0.0036 -0.8095 0.4182 -0.0099 0.0041
## practice_phaseyes 0.1027 0.3093 0.3320 0.7399 -0.5035 0.7088
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = test_mass_or_distributed)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by test type")
m_age_tt <- rma.mv(d_calc ~ mean_age + test_mass_or_distributed, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_tt)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -55.3701 110.7401 118.7401 120.6797 124.4544
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.1332 1.0645 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 12) = 148.1928, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 6.9802, p-val = 0.0305
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt -3.5605 1.8703 -1.9037 0.0569 -7.2262
## mean_age 0.0035 0.0014 2.5647 0.0103 0.0008
## test_mass_or_distributedmass 2.5529 1.3042 1.9575 0.0503 -0.0032
## ci.ub
## intrcpt 0.1051 .
## mean_age 0.0062 *
## test_mass_or_distributedmass 5.1090 .
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_tt_interaction <- rma.mv(d_calc ~ mean_age * test_mass_or_distributed, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data)
summary(m_age_tt_interaction)
##
## Multivariate Meta-Analysis Model (k = 15; method: REML)
##
## logLik Deviance AIC BIC AICc
## -53.2590 106.5180 116.5180 118.5075 128.5180
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 1.8313 1.3533 4 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 11) = 147.2843, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 10.4552, p-val = 0.0151
##
## Model Results:
##
## estimate se zval pval
## intrcpt -5.3020 2.1983 -2.4119 0.0159
## mean_age 0.0049 0.0016 3.0994 0.0019
## test_mass_or_distributedmass 8.7583 3.4873 2.5115 0.0120
## mean_age:test_mass_or_distributedmass -0.0077 0.0039 -1.9699 0.0488
## ci.lb ci.ub
## intrcpt -9.6106 -0.9935 *
## mean_age 0.0018 0.0080 **
## test_mass_or_distributedmass 1.9233 15.5933 *
## mean_age:test_mass_or_distributedmass -0.0153 -0.0000 *
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_young %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = test_mass_or_distributed)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by test_mass_or_distributed, young only")
m_age_tt_young <- rma.mv(d_calc ~ mean_age + test_mass_or_distributed, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_tt_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.3250 56.6500 64.6500 64.4337 84.6500
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0997 1.4490 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 7) = 100.0259, p-val < .0001
##
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 1.4077, p-val = 0.4947
##
## Model Results:
##
## estimate se zval pval ci.lb
## intrcpt 2.5154 3.5290 0.7128 0.4760 -4.4013
## mean_age -0.0028 0.0035 -0.8014 0.4229 -0.0097
## test_mass_or_distributedmass 1.0084 1.9222 0.5246 0.5999 -2.7590
## ci.ub
## intrcpt 9.4321
## mean_age 0.0041
## test_mass_or_distributedmass 4.7757
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
m_age_tt_interaction_young <- rma.mv(d_calc ~ mean_age * test_mass_or_distributed, V = d_var_calc,
random = ~ 1 | short_cite, data = ma_data_young)
summary(m_age_tt_interaction_young)
##
## Multivariate Meta-Analysis Model (k = 10; method: REML)
##
## logLik Deviance AIC BIC AICc
## -28.3118 56.6236 66.6236 65.5824 126.6236
##
## Variance Components:
##
## estim sqrt nlvls fixed factor
## sigma^2 2.0957 1.4476 3 no short_cite
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 99.8371, p-val < .0001
##
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 1.4401, p-val = 0.6962
##
## Model Results:
##
## estimate se zval pval
## intrcpt 7.7218 29.7396 0.2596 0.7951
## mean_age -0.0085 0.0326 -0.2619 0.7934
## test_mass_or_distributedmass -4.2465 29.8649 -0.1422 0.8869
## mean_age:test_mass_or_distributedmass 0.0058 0.0328 0.1763 0.8600
## ci.lb ci.ub
## intrcpt -50.5668 66.0105
## mean_age -0.0725 0.0554
## test_mass_or_distributedmass -62.7806 54.2875
## mean_age:test_mass_or_distributedmass -0.0585 0.0701
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ma_data_old %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = age_months, y = d_calc, size = n_1, color = test_mass_or_distributed)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("Age (months)") +
ggtitle("Syntactical Bootstrapping effect size vs. Age (months), breakdown by test_mass_or_distributed, old only")
ma_data_vocab %>%
mutate(age_months = mean_age/30.44) %>%
ggplot(aes(x = productive_vocab_median, y = d_calc, size = n_1, color = test_mass_or_distributed)) +
geom_point() +
geom_smooth(method = "lm") +
ylab("Effect Size") +
xlab("median vocab") +
ggtitle("Syntactical Bootstrapping effect size vs. median voca, breakdown by test_mass_or_distributed, old only")
ma_data %>% ggplot(aes(x = n_train_test_pair)) +
geom_histogram()
ma_data %>% ggplot(aes(x = n_test_trial_per_pair)) +
geom_histogram()